CD28-mediated cytotoxicity of YT natural killer cells on B7-positive targets induces rapid necrotic death independent of granule exocytosis

Granzyme A from cytotoxic lymphocytes cleaves GSDMB to trigger pyroptosis in target cells

Cytotoxic lymphocyte-mediated immunity relies on granzymes. Granzymes are thought to kill target cells by inducing apoptosis, although the underlying mechanisms are not fully understood. Here, we report that natural killer cells and cytotoxic T lymphocytes kill gasdermin B (GSDMB)-positive cells through pyroptosis, a form of proinflammatory cell death executed by the gasdermin family of pore-forming proteins. Killing results from the cleavage of GSDMB by lymphocyte-derived granzyme A (GZMA), which unleashes its pore-forming activity. Interferon-γ (IFN-γ) up-regulates GSDMB expression and promotes pyroptosis. GSDMB is highly expressed in certain tissues, particularly digestive tract epithelia, including derived tumors. Introducing GZMA-cleavable GSDMB into mouse cancer cells promotes tumor clearance in mice. This study establishes gasdermin-mediated pyroptosis as a cytotoxic lymphocyte-killing mechanism, which may enhance antitumor immunity.

Substance P inhibits natural killer cell cytotoxicity through the neurokinin-1 receptor

SP is a potent neuroimmunomodulator that functions through ligating members of the neurokinin receptor family, one of which, NK1R, is widely expressed in immune cells. As in humans, circulating SP levels are increased in pathologic states associated with impairment of NK cell functions, such as depression and HIV infection, we hypothesized that SP has a direct, inhibitory effect upon NK cells. We have studied a clonal human NK cell line (YTS) as well as ex vivo human NK cells and have determined that truncated and full-length NK1R isoforms are expressed in and SP bound by ex vivo NK cells and the YTS NK cell line. Incubation of YTS cells with 10⁻⁶ M SP and ex vivo NK cells with 10⁻⁵ M SP inhibited cytotoxic ability by ∼20% and reduced degranulation. This inhibitory effect upon cytotoxicity was partially prevented by the NK1R antagonist CP96,345. The treatment of YTS or ex vivo NK cells with SP neither down-modulated NCR expression nor affected triggering receptor-induced NF-κB activation. Preincubation of YTS cells with SP, however, did abbreviate the typically prolonged intracellular calcium increase induced by target cell engagement and reduced triggering receptor-induced pERK. Thus, SP has the potential to regulate NK cell functions and acts downstream from neurokinin receptors to modulate NK cell activation signaling. This mechanism may contribute to impairment of NK cell function in certain disease states associated with increased circulating SP. Antagonism of this system may present an opportunity to augment NK cell function therapeutically in selected human diseases.

Spontaneous clustering and tyrosine phosphorylation of NK cell inhibitory receptor induced by ligand binding

Inhibition of NK cell cytotoxicity by killer cell Ig-like receptors (KIR) depends on phosphorylation of cytoplasmic tyrosines in KIR, which recruit tyrosine phosphatase Src homology protein tyrosine phosphatase 1. It is not clear how KIR, whose function lies downstream of a tyrosine kinase, succeeds in blocking proximal NK cell activation signals upon binding HLA class I on target cells. Here we show that mixing NK cells with insect cells expressing HLA-C was sufficient to induce clustering of KIR, and phosphorylation of KIR and SHP-1. Transient phosphorylation of KIR was detected in the presence of pervanadate, an inhibitor of protein tyrosine phosphatases, at suboptimal concentration. Phosphorylation of KIR was specifically induced by ligand binding because it was detected only when HLA-C was loaded with a peptide that permits KIR binding. KIR phosphorylation was not dependent on ICAM-1-mediated adhesion and was not blocked by inhibition of actin polymerization, but required Zn(2+). Fluorescence resonance energy transfer between HLA-C molecules revealed close molecular interactions induced by KIR binding. These results demonstrate tight clustering of KIR and rapid KIR phosphorylation induced simply by binding to HLA-C. The unique property of KIR to become phosphorylated in the absence of adhesion and of actin cytoskeleton rearrangement explains how KIR can efficiently block early activation signals during NK-target cell contacts.

Targeting of Human Dendritic Cells by Autologous NK Cells

NK cells have the capacity to spontaneously kill tumor cell lines, in particular cell lines of hemopoietic origin. In contrast, they do not generally kill nontransformed autologous cells. However, here we demonstrate that short-term activated polyclonal human NK cells, as well as human NK cell lines, efficiently lyse autologous dendritic cells (DC) derived from peripheral blood monocytes as well as Langerhans-like cells derived from CD34+ stem cells isolated from umbilical cord blood. Lysis of autologous DC by short-term activated NK cells and NK cell lines was dependent on granule exocytosis, since total abrogation of lysis was observed in the presence of EGTA. Induction of DC maturation by LPS, monocyte conditioned media (MCM), or stimulation through CD40 ligand (CD40L) rendered the DC less susceptible to lysis by NK cells. Infection of DC with influenza virus was likewise associated with a reduced susceptibility to lysis by NK cells. Thus, susceptibility to lysis by autologous NK cells is a particular property of immature DC. The present results are discussed in relation to the ability of DC to interact with NK cells and to the ability of NK cells to regulate development of specific immunity.

NK Cell Triggering by the Human Costimulatory Molecules CD80 and CD86

NK cell-mediated effector functions are regulated by a delicate balance between positive and negative signals. Receptors transmitting negative signals upon engagement with target cell MHC class I molecules have been characterized in detail in recent years. In contrast, less information is available about receptor-ligand interactions involved in the transmission of positive or “triggering” signals to NK cells. Recently, it has been described that murine NK cells are triggered by the costimulatory molecules CD80, CD86, and CD40. Using NK cell lines derived from PBMC as effectors, we demonstrate that the human CD80 and CD86 gene products can function as triggering molecules for NK cell-mediated cytotoxicity. Expression of human CD80 or CD86 molecules in murine B16.F1 melanoma cells rendered these significantly more susceptible to lysis by human NK cell lines. Blocking of the transfected gene products with specific mAb reduced lysis levels to that of nontransfected control cell lines. Triggering of human NK cells by CD80 and CD86 appeared to be independent of CD28 and CTLA-4, at least as determined by the reagents used in the present study, because the expression of these molecules could not be detected on the NK cell lines by either flow cytometry or in redirected lysis assays. Thus, human NK cells may use receptors other than CD28 and CTLA-4 in their interactions with CD80 and CD86 molecules. Alternatively, interactions may involve variants of CD28 (and possibly CTLA-4) that are not recognized by certain anti-CD28 mAb.

Expression of a Variant of CD28 on a Subpopulation of Human NK Cells: Implications for B7-Mediated Stimulation of NK Cells

The ability of NK cells to kill tumor cells is controlled by a balance between activating and inhibitory signals transduced by distinct receptors. In murine tumor models, the costimulatory molecule B7.1 not only acts as a positive trigger for NK-mediated cytotoxicity but can also overcome negative signaling transduced by MHC class I molecules. In this study, we have evaluated the potential of human B7.1-CD28 interaction as an activating trigger for human blood NK cells. Using multiparameter flow cytometric analysis and a panel of different CD28 mAbs, we show that human peripheral blood NK cells (defined by CD56+, CD16+, and CD3− surface expression) express the CD28 costimulatory receptor, with its detection totally dependent on the mAb used. In addition, the level of CD28 varies among individuals and on different NK cell lines, irrespective of CD28 steady-state mRNA levels. By performing Ab binding studies on T cells, our data strongly suggest that binding of two of the anti-CD28 Abs (clones 9.3 and CD28.2) is to a different epitope to that recognized by clones L293 and YTH913.12, which is perhaps modified in the CD28 molecule expressed by the NK cells. We also show that B7.1 enhances the NK-mediated lysis of NK-sensitive but not of NK-resistant tumor cells and that this increased lysis is dependent on CD28-B7 interactions as shown by the ability of Abs to block this lysis. Coculture of the B7.1-positive NK-sensitive cells also led to the activation of the NK cells, as determined by the expression of CD69, CD25, and HLA class II.

Interaction between B.7 and CD28 costimulatory molecules is essential for the activation of effector function mediating spontaneous tumour regression

The spontaneous regression of a rat histiocytoma, AK-5, is mediated by activated natural killer cells through antibody-dependent cellular cytotoxicity. In addition to the Fc-FcR interaction between the target and the effector cells demonstrated previously, we show the participation of costimulatory molecules B7 and CD28 in the efficient killing of the tumour cell. Blockade of the costimulatory interaction in vivo using anti-CD28 led to increased tumour growth and a suppressed cytokine response. Anti-CD28 antibody administration in vivo also diminished the cytotoxic potential of NK cells against AK-5 cells in vitro. Our studies also demonstrate the expression of B7.1 and B7.2 antigen on AK-5 tumour cells. The cytotoxic activity of natural killer cells was significantly inhibited when the effector/target cells were cultured in the presence of antibodies raised against B7.1, B7.2 and CD28. Administration of anti-CD28 in vivo also affected the efficiency of the formation of effector/target conjugates in vitro. Similarly, anti-CD28 injections affected expression of the adhesion molecules LFA 1 and ICAM 1 by splenocytes. Administration of anti-B7.1 and B7. 2 antibodies in AK-5 tumour-bearing animals showed a differential response. The cytotoxicity of natural killer cells was significantly inhibited after anti-B7.2 administration, suggesting the preferential participation of B7.2 molecules in vivo. These observations suggest an important role for B7-CD28 interaction in AK-5 tumour regression.

Pit cells (Hepatic natural killer cells) of the rat induce apoptosis in colon carcinoma cells by the perforin/granzyme pathway

The high mortality of colon cancer is to a large extent caused by the frequent occurrence of liver metastasis. This is remarkable, because the liver harbors two distinct cell populations that can eliminate invading cancer cells, namely hepatic natural killer (NK) cells and Kupffer cells. These hepatic NK cells, known as pit cells, are the most cytotoxic cells of the naturally occurring NK cells. However, the mechanism by which pit cells eliminate tumor cells is largely unknown. Because we recently found an indication that apoptosis is involved, we tried to assess the role of this mode of cell death using an in vitro system with isolated pure pit cells (>90%) and CC531s cells, a rat colon carcinoma (CC) cell line. Pit cells induced apoptosis in CC531s cells as shown by quantitative DNA fragmentation, agarose gel electrophoresis, and different modes of microscopy. When extracellular Ca2+ was chelated by ethylene glycol-bis(beta-aminoethyl ether)-N,N,-tetraacetic acid (EGTA) during coincubation or when the pit cells were preincubated with the granzyme inhibitor 3,4-dichloroisocoumarin (DCI), the induction of apoptosis was abolished. These results show that pit cells use the Ca2+-dependent perforin/granzyme pathway to induce apoptosis in the CC531s cells, and not the alternative Ca2+-independent Fas pathway. To further exclude the possibility of the involvement of the Fas pathway, we treated CC531s cells with recombinant Fas ligand. This treatment did not result in the induction of apoptosis, indicating that CC531s cells are resistant to Fas-mediated apoptosis. We conclude therefore that pit cells induce apoptosis in CC cells in vitro by the perforin/granzyme pathway.

Antigen-dependent CD28 signaling selectively enhances survival and proliferation in genetically modified activated human primary T lymphocytes

Most tumor cells function poorly as antigen-presenting cells in part because they do not express costimulatory molecules. To provide costimulation to T lymphocytes that recognize tumor cells, we constructed a CD28-like receptor specific for GD2, a ganglioside overexpressed on the surface of neuroblastoma, small-cell lung carcinoma, melanoma, and other human tumors. Recognition of GD2 was provided by a single-chain antibody derived from the GD2-specific monoclonal antibody 3G6. We demonstrate that the chimeric receptor 3G6-CD28 provides CD28 signaling upon specific recognition of the GD2 antigen on tumor cells. Human primary T lymphocytes retrovirally transduced with 3G6-CD28 secrete interleukin 2, survive proapoptotic culture conditions, and selectively undergo clonal expansion in the presence of an antiidiotypic antibody specific for 3G6-CD28. Polyclonal CD8(+) lymphocytes expressing 3G6-CD28 are selectively expanded when cultured with cells expressing allogeneic major histocompatibility complex class I together with GD2. Primary T cells given such an antigen-dependent survival advantage should be very useful to augment immune responses against tumor cells.

Natural killer cell reactivity: activation and cytolysis mechanism models, involving heat shock protein, haemopoietic histocompatibility, major histocompatibility complex and complement molecules

The close association of heat shock protein (HSP), haemopoietic histocompatibility (Hh), major histocompatibility complex (MHC), and complement genes on the same chromosomal region, and the fact that all these genes are inherited on the whole in each haplotype of an individual, might indicate some evolutionary and functional correlations among them. Several data suggest for HSP70 molecules a possible role as a molecular target recognizable by natural killer (NK) cells. HSP70 sequences from both prokaryotic and eukaryotic organisms reveal that about half of the amino acid residues are identical and many of the remaining residues are similar. I here assume that NK reactivity might start, early in the immunogenesis process, as a effect of the interaction between HSP70 molecules and a hypothetical HSP receptor of yet immature non-cytolytic NK cells. To this receptor, an HSP molecule might act as an activator or an inhibitor depending on whether its amino acid residues are reactive or not with it, respectively. Later in the immunogenesis process, murine Hh or human equivalent molecules, dominantly expressed in bone marrow target cells, might select the non-reactive NK clones of an individual, inducing them to mature and express a lytic machinery. As a consequence of the NK maturation, proliferating hemopoietic target cells expressing only or mainly activator HSPs on their surface might undergo NK cytolysis. This might explain the NK lysis of apparently normal cells found in human foetal marrow; moreover, this might explain in some way the F1 hybrid resistance phenomenon. The NK reactivity of an individual would be further modulated by the expression on the NK surface of particular receptors (CD94, p58) specific for defined MHC molecules (Cw1, Cw3, Bw6, B7) on the target cells. Such a specific interaction would induce an 'NK effector inhibition'. The NK reactivity mechanism might have been further evolutionarily modified and adapted by the involvement of other NK receptors, such as CD11b (specific for the C3b factor of the complement) and CD16 (specific for the IgG Fc piece). Cooperation among HSP, MHC, CD11b, CD16, C3b and Fc allows us to propose original models of the activation and cytolysis mechanisms in the NK cytotoxicity and antibody-dependent cell cytotoxicity phenomena.

Absence of B7.1-CD28/CTLA-4-mediated co-stimulation in human NK cells

Recent studies have suggested that B7-CD28 interactions provide co-stimulatory signals for activation of NK cells. Transduction of the B7.1 (CD80) gene into tumor cells has been shown to trigger proliferation and cytotoxicity of murine NK cells and a human NK cell line, YT2C2. Therefore, transduction of the B7.1 gene into CD80-negative human squamous cell carcinomas of the head and neck (SCCHN) and its stable expression was expected to upregulate proliferation and cytotoxic activities of human NK cells. However, expression of the B7.1 receptors, CD28 and CTLA-4, could not be demonstrated on the surface or in the cytoplasm of normal human NK cells, irrespective of the state of their activation. In proliferation experiments or various cytotoxicity assays, utilizing highly purified human NK cells as responder or effector cells, no enhancement of NK cell generation or activity, respectively, by B7.1+ SCCHN was observed relative to non-transduced or LacZ gene-transduced SCCHN. In contrast, co-incubation of B7.1+ SCCHN targets with human NK cells induced significant inhibition of NK cell growth. Thus, the B7.1-CD28/CTLA-4 pathway is not involved in triggering of human adult NK cells.

Calcium ionophore-induced degradation of neurofilament and cell death in MSN neuroblastoma cells

Extensive necrotic death of MSN neuroblastoma cells could be induced after incubation with the calcium ionophore, A23187. The reaction was concentration-dependent and time course-dependent. Levels of the 66 kd/alpha-internexin neurofilament protein (NF-66) and the cognate heat shock protein 70 (Hsc 70) decreased during the Ca2+-activated cell death. Addition of the calcium chelator, ethylene glycol-bis(beta-aminoethyl ether) N,N,N',N'-tetraacetic acid (EGTA) restored the normal level of NF-66 and partially that of the Hsc 70. Use of either calpain I or calpain II inhibitor could alleviate the reduction of 66 kd protein during the ionophore treatment whereas only calpain I inhibitor treatment was effective in restoring the normal level of the Hsc 70. Neither of these calpain inhibitors could block the ionophore triggered cell death. EGTA was toxic to cells in a wide range of concentration suggesting a calcium-independent activation of cell death mechanism.

Triggering of natural killer cells by the costimulatory molecule CD80 (B7-1)

NK cell-mediated cytotoxicity is influenced by triggering as well as inhibitory signals. The identification of inhibitory signals provided by MHC class I molecules has recently attracted significant attention. Much less is known about putative triggering signals. Using purified populations of mouse NK cells, we demonstrate that the CD80 (B7-1) gene product functions as a triggering signal for NK cell-mediated cytotoxicity. The strength of this response is such that it overrides the protection mediated by MHC class I molecules. Triggering of mouse NK cells by B7-1 occurred even in the absence of CD28 and could not be blocked by either anti-CD28 or anti-CTLA-4 antibodies. NK cells may thus, at least in part, use receptors other than CD28 and CTLA-4 in their interaction with B7-1. Furthermore, we demonstrate that bone marrow-derived macrophages and dendritic cells are highly susceptible to lysis by autologous NK cells.